Inhibition of DNA synthesis by a deoxynucleoside or a deoxynucleoside analogue can kill cells ("direct toxicity") and can also enhance the toxicity of alkylating agents ("sensitization"). Not surprisingly, the extent of direct toxicity correlates closely with the degree of sensitization. This proposal addresses T cell neoplasms which are unusually sensitive to direct toxicity by deoxyguanosine and uniquely and exquisitely sensitive to its analogue 9-beta-D-arabinofuranosylguanine ("araG"). One objective of this proposal is to use this unique sensitivity to selectively sensitize T cell neoplasms compared to normal hematopoietic progenitor cells responsible for dose-limiting clinical toxicity. Such T cell selective sensitization has clear potential applications in clinical treatment as well as in ex vivo purging of neoplastic T cells from bone marrow. A second objective is to determine the mechanisms of sensitization with the long term intention of extrapolating these mechanistic concepts to closely analogous sensitization approaches. In preliminary studies araG and deoxyguanosine did potently enhance alkylator toxicity producing 5 to 8 fold dose modifying effects in human T cell neoplasms but not in B cells. In sharp contrast, guanosine was non- toxic and strongly protected against alkylator toxicity in human T cell lines - despite depleting dCTP and dTTP pools to the same extent as did equimolar deoxyguanosine. Guanosine's surprising effect and the very different effect of araG may relate to metabolism of araG almost exclusively via the deoxycytidine kinase ("dCK") pathway and to metabolism of guanosine via the purine nucleoside phophorylase ("PNP") pathway. Deoxyguanosine, metabolized via both pathways, might sensitize T cell neoplasms (high dCK:PNP ratio) while simultaneously protecting non-T cells (low dCK:PNP). Using measures of nucleoside metabolism, DNA injury and repair, and new techniques for specifically identifying which aspects of DNA synthesis are being impaired, this proposal will test l) whether deoxyguanosine, araG or guanosine preferentially sensitize or protect human T cell malignancies relative to non-T cells including normal marrow progenitor cells, 2) whether deoxyguanosine can sensitize or protect depending on its metabolism via the dCK or PNP pathways, 3) whether guanosine protects by increasing guanylate ribonucleotides which then block cell cycle progression permitting greater repair of alkylator- injured DNA before cells initiate replicative synthesis, and 4) whether deoxyguanosine and araG cause direct toxicity and sensitization not only by depleting dCTP and dTTP but also by high dGTP and araGTP inhibiting DNA polymerases alpha, delta or epsilon.